Gram negative infections are a major cause of morbidity and mortality especially in hospitalized and immunocompromised patients. [Duma, R. J., Am. J. of Med., 78 (Suppl. 6A): 154-164 (1985); and Kreger B. E., D. E. Craven and W. R. McCabe, Am. J. Med., 68: 344-355 (1980)] Antibiotics are presently being used to contain infections.
LPS is a major component of the outer membrane of gram negative bacteria and is released when the organisms are lysed. [Shenep, J. L. and K. A. Morgan, J. Infect. Dis., 150 (3): 380-388 (1984)] Although available antibiotics are generally effective in containing the infection, they do nothing to neutralize the pathophysiological effects associated with lipopolysaccharide (LPS).
LPS released during antibiotic therapy is a potent stimulator of the inflammatory response. Many detrimental effects of LPS in vivo result from soluble mediators released by inflammatory cells. [Morrison D. C. and R. J. Ulevich, Am. J. Pathol., 93 (2): 527-617 (1978)] LPS induces the release of mediators by host inflammatory cells which may ultimately result in disseminated intravascular coagulation (DIC), adult respiratory distress syndrome (ARDS), renal failure, and irreversible shock.
Soluble LPS causes decreased neutrophil chemotaxis, increased adhesiveness, elevated hexose monophosphate shunt activity and O.sub.2 radical production, upregulation of surface receptors for complement, and release of granule proteins into the surrounding medium. [Morrison and Ulevich (1978)]
Endotoxemia is a condition associated with the presence of endotoxins, i.e. heat stable bacterial toxins, in the blood. Endotoxins elicit an inflammatory response that is beneficial in fighting the infection but can be damaging to the host if uncontrolled. Endotoxemia induces production of LPS binding proteins from the liver and causes release of microbicidal proteins from leukocytes. Our studies show that one of these leukocytes proteins (BPI) previously known only for its bactericidal activity in vitro, inhibits the ability of LPS to stimulate neutrophils and monocytes and reduces death due to endotoxin or bacterial challenge when given in vivo.
Monocytes and neutrophilic granulocytes play a key role in host defense against bacterial infections and also participate in the pathology of endotoxemia. These cells ingest and kill microorganisms intracellularly and also respond to LPS in vivo and in vitro by releasing soluble proteins with microbicidal, proteolytic, opsonic, pyrogenic, complement activating and tissue damaging effects.
Tumor necrosis factor (TNF), a cytokine released by LPS stimulated monocytes mimics some of the toxic effects of LPS in vivo. Injecting animals with TNF causes fever, shock and alterations in glucose metabolism. TNF is also a potent stimulator of neutrophils.
Despite improvements in antibiotic therapy, morbidity and mortality associated with endotoxemia remains high. Antibiotics alone are not effective in neutralizing the toxic effects of LPS. Therefore, the need arises for an adjunct therapy with direct LPS neutralizing activity. Current methods for treatment of endotoxemia use antibiotics and supportive care. Most available adjunct therapies treat symptoms of endotoxic shock such as low blood pressure and fever but do not inactivate endotoxin. Other therapies inhibit inflammatory host responses to LPS. As indicated below, present therapies have major limitations due to toxicity, immunogenicity, or irreproducible efficacy between animal models and human trials.
PMB is a basic polypeptide antibiotic which has been shown to bind to, and structurally disrupt, the most toxic and biologically active component of endotoxin, Lipid A. PMB has been shown to inhibit LPS activation of neutrophil granule release in vitro and is an effective treatment for gram negative sepsis in humans. However, because of its systemic toxicity, this drug has limited use except as a topical agent.
Combination therapy using antibiotics and high doses of methylprednisolone sodium succinate (MPSS) has been shown to prevent death in an experimental model of gram negative sepsis using dogs. Another study using MPSS with antibiotics in a multicenter, double blind, placebo-controlled, clinical study in 223 patients with clinical signs of systemic sepsis concluded that mortality was not significantly different between the treatment and placebo groups. Further, the investigators found that resolution of secondary infection within 14 days was significantly higher in the placebo group.
A relatively new approach to treatment of endotoxemia is passive immunization with endotoxin neutralizing antibodies. Hyperimmune human immunoglobulin against E. coli J5 has been shown to reduce mortality in patients with gram negative bacteremia and shock by 50%. Other groups have shown promising results in animal models using mouse, chimeric, and human monoclonal antibodies. Although monoclonal antibodies have advantages over hyperimmune sera, e.g. more consistent drug potency and decreased transmission of human pathogens, there are still many problems associated with administering immunoglobulin to neutralize LPS. Host responses to the immunoglobulins themselves can result in hypersensitivity. Tissue damage following complement activation and deposition of immune complexes is another concern in the use of therapies involving anti-endotoxin antibodies in septic patients.
BPI was first discovered in 1975 [Weiss, J., R. C. Franson, S. Becherdite, K. Schmeidler, and P. Elsbach, J. Clin. Invest., 55:33 (1975)] and was first obtained in highly purified form from human neutrophils in 1978 and shown to be bactericidal against gram negative bacteria when assayed in phosphate buffered saline in vitro [Weiss, J., P. Elsbach, I. Olson and H. Odeberg, J. Biol. Chem, 253 (8): 2664-2672 (1978)]. The mechanism of bacterial killing was not defined but proposed to be mediated through changes in membrane permeability (thus the name bactericidal/permeability increasing protein). However, BPI is inhibited by Mg, Ca, heparin [Weiss et al. J. Biol. Chem., 253 (8): 2664-2672 (1978)], and serum albumin [Mannion et al. J. Clin. Invest. 85: 853-860 (1990)] suggesting that it is not bactericidal in vivo. Weiss et al. [J. Biol. Chem. 254 (21): 11010-11014 (1979)], further showed that BPI increased phospholipase A2 activity suggesting a proinflammatory activity for BPI in addition to its supposed bactericidal activity.
Rabbit BPI was purified in 1979 [Elsbach et al. J. Biol. Chem. 254 (21): 11000-11009] and shown to have identical bactericidal and permeability increasing properties as human BPI providing a further source of material for study. Both rabbit and human BPI were shown to be effective against a variety of gram negative bacteria in vitro, including K1-encapsulated E. coli [Weiss et al. Infection and Immunity 38 (3): 1149-1153, (1982)].
A role for lipopolysaccharide in the in vitro bactericidal action of BPI was proposed in 1984 by Weiss et al. [J. Immunol. 132 (6): 3109-3115, (1984)] who demonstrated that BPI bound to the outer membrane of gram-negative bacteria and caused extracellular release of LPS and selectively stimulated biosynthesis of LPS. In 1984 a 57 kD protein with similar properties was isolated from human neutrophils and designated CAP 57 [Shafer, W. M., C. E. Martin and J. K. Spitznagel, Infect. Immun., 45:29 (1984)] This protein is identical to BPI protein as determined by the N-Terminal amino acid sequence, amino acid composition, molecular weight and source [Spitznagel et al Blood 76:825-834, 1990]. Another group, Hovde and Gray reported a 55kDa bactericidal glycoprotein with virtually identical properties to BPI in 1986 [Hovde and Gray Infection and Immunity 54(1): 142-148 (1986)].
BPI retains its in vitro bactericidal activity after cleavage of BPI with neutrophil proteases suggesting that fragments of the molecule retain activity [Ooi and Elsbach Clinical Research 33 (2):567A, (1985). All of the in vitro bactericidal and permeability increasing activities of BPI were later shown to be present in the N-terminal 25 kD fragment of the protein. [Ooi, C. E., J. Weiss, P. Elsbach, B. Frangione, and B. Marrion, J. Biol. Chem., 262: 14891 (1987)]
The fact that BPI is an LPS binding protein is evidenced by: (1) increased sensitivity of rough strains of permeability increasing activities of BPI [Weiss, J., M. Hutzler and L. Kao, Infect. Immun., 51:594 (1986)]; (2) mutations in the Lipid A domain of LPS caused decreased binding and increased resistance to bactericidal activity of both polymyxin B and BPI [Farley, M. M., W. M. Shafer and J. K. Spitznagel, Infect. Immun., 56:1536-1539 (1987) and Farley et al. Infect. Immun. 58:1589-1592 (1988)]; (3) BPI competed with polymyxin B (PMB) for binding to S. typhimurium [Farley 1988]; (4) BPI protein sequence homology and immunocrossreactivity to another LPS binding protein termed Lipopolysaccharide Binding Protein (LBP) [Tobias et al. J. Biol. Chem. 263 (27): 13479-13481 (1988). LBP-LPS complexes have been shown to stimulate the oxidative burst of neutrophils in response to formylated peptides [Vosbeck et al. Eur. J. Clin. Invest. 18 A50 (1988)]. In addition, LBP-LPS complexes bind back to a cell surface receptor on monocytes (CD 14) resulting in the induction of tumor necrosis factor (TNF) [Schumann et al. Science 249:1429-1431]. Thus LBP mediates the immunostimulatory activity of LPS and therefore stimulates the toxic response to endotoxin. BPI has exactly the opposite effects of LBP, binding to LPS and inhibiting neutrophil activation and blocking TNF production by monocytes.
BPI binding to gram negative bacteria was reported originally to disrupt LPS structure, alter microbial permeability to small hydrophobic molecules and cause cell death (Weiss, et al., 1978). More recently these same authors have demonstrated that such effects occur only in the absence of serum albumin. If bacteria are cultured in the presence of serum albumin BPI, in fact, has no bactericidal activity thus proving that BPI does not kill bacteria in vivo [Mannion et Clin. Invest. 85: 853-860 (1990) and Mannion et al. J. Clin. Invest. 86: 631-641)]. Therefore, prior to the subject invention, it has been understood in the art that the beneficial effects of BPI protein are limited to in vitro bactericidal effects. Here we show that BPI protein binds endotoxin in the presence of serum and plasma and, unlike other known LPS binding proteins such as LBP, BPI inhibits the immunostimulatory and toxic activities of LPS both in vitro and in vivo respectively. Thus BPI has a novel and distinct use in the therapeutic and prophylactic treatment of endotoxin-related disorders including endotoxemia and endotoxic shock.
Furthermore BPI is described by Gray et al. [J. Biol. Chem. 264 (16) 9505-9509 (1989)] as a membrane protein which must be cleaved to the 25 kDa fragment to be released from the neutrophil granule membrane in soluble form. The present invention provides for a method of producing full length soluble BPI in active form. Further the present invention separates for the first time two molecular forms of the molecule apparantly unresolved by Gray et al. representing glycosylated and nonglycosylated forms of the molecule which appear to have different serum half-life profiles in vivo and thus different therapeutic potential. Natural BPI from neutrophils is a mixture of the glycoslyated and nonglycosylated forms.